CN102902587A - Distribution type task scheduling method, distribution type task scheduling system and distribution type task scheduling device - Google Patents
Distribution type task scheduling method, distribution type task scheduling system and distribution type task scheduling device Download PDFInfo
- Publication number
- CN102902587A CN102902587A CN2011102144282A CN201110214428A CN102902587A CN 102902587 A CN102902587 A CN 102902587A CN 2011102144282 A CN2011102144282 A CN 2011102144282A CN 201110214428 A CN201110214428 A CN 201110214428A CN 102902587 A CN102902587 A CN 102902587A
- Authority
- CN
- China
- Prior art keywords
- task
- groove
- resource
- scheduling controller
- tasks
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Multi Processors (AREA)
Abstract
The invention discloses a distribution type task scheduling method, a distribution type task scheduling system and a distribution type task scheduling device. The distribution type task scheduling system comprises a control server and a plurality of execution servers, wherein the control server comprises a scheduling controller; each execution controller comprises a plurality of task grooves; a certain amount of resources are assigned to each task groove; when the system is operated, the scheduling controller maintains two queues including a task waiting queue and a task groove queue; when each task X is received, if the task groove in the task groove queue is suitable for processing the task X, the task X is assigned to the task groove; otherwise, the task X is added into the task waiting queue; when a task request actively sent by the task groove X is received for each time, if a task suitable for being processed by the task groove X exists in the task waiting queue, the task is assigned to the task groove X; and otherwise, the task groove X is added into the task groove queue. With the adoption of the scheme provided by the invention, the resource use ratio is increased and the task processing efficiency can be improved.
Description
Technical field
The present invention relates to the task scheduling technology, particularly the distributed task dispatching mthods, systems and devices.
Background technology
Distributed task dispatching refers to one is used or a plurality of tasks that delineation of activities goes out are transferred to a plurality of servers and processed.Distributed task dispatching can solve the problems such as individual server inadequate resource, reaches the purpose that large-scale data calculates.
Usually, can comprise a Control Server in the distributed task dispatching system, other each server is called the execution server, can utilize Control Server to come that each is carried out server and carry out unified management, as when having task to carry out, according to certain rule it is distributed to a certain execution server by Control Server.
For carrying out server, it can only accept the task of distributing passively, but this task may and be not suitable for its processing, such as: the free memory of supposing the execution server is 700M, the task of current execution has taken 200M, a newly assigned task need to take the internal memory of 600M, so, this newly assigned task just can't be carried out, and namely can only wait until after current task is finished and could carry out, and so remaining 500M internal memory just is equal to and has been wasted, and, if current task need to be carried out the long time, newly assigned task just needs to wait for the long time, thereby has reduced the task treatment effeciency.
Summary of the invention
In view of this, fundamental purpose of the present invention is to provide a kind of distributed task dispatching method, can improve resource utilization and task treatment effeciency.
Another object of the present invention is to provide a kind of distributed task dispatching system, can improve resource utilization and task treatment effeciency.
Another purpose of the present invention is to provide a kind of execution server, can improve resource utilization and task treatment effeciency.
A further object of the present invention is to provide a kind of Control Server, can improve resource utilization and task treatment effeciency.
For achieving the above object, technical scheme of the present invention is achieved in that
A kind of distributed task dispatching method, a Control Server is set in the distributed task dispatching system and carries out server more than one, a scheduling controller is set in Control Server, in each execution server, mark off respectively two above task grooves, and be that each task groove distributes required resource; When the operation of described distributed task dispatching system, the method comprises:
Scheduling controller is safeguarded two formations, and one is the task waiting list, and another is the task slot queue;
When whenever receiving a task X, whether scheduling controller sets the tasks exists the task groove that is fit to Processing tasks X in the slot queue, if so, then task X is distributed to this task groove, otherwise, task X is joined in the task waiting list;
When whenever receiving the task requests that a task groove X initiatively sends, whether scheduling controller sets the tasks to exist in the waiting list and is fit to the task that task groove X processes, and if so, then this task is distributed to task groove X, otherwise, task groove X is joined in the task slot queue; Described task requests is that task groove X determines that the own resource operating position is fit to send after the new task of active request.
A kind of distributed task dispatching system comprises: two above servers, and one of them is Control Server, or is Control Server simultaneously and carries out server, all the other are the execution server; Comprise a scheduling controller in the Control Server, each is carried out in the server and includes two above task grooves, and each task groove all is assigned certain resource, further comprises telegon and tasks carrying device in each task groove;
Scheduling controller is used for safeguarding two formations that one is the task waiting list, and another is the task slot queue; When whenever receiving a task X, whether there is the task groove that is fit to Processing tasks X in the slot queue that sets the tasks, if so, then task X is distributed to this task groove, otherwise, task X is joined in the task waiting list; When whenever receiving the task requests that a task groove X initiatively sends, whether exist in the waiting list that sets the tasks to be fit to the task that task groove X processes, if so, then this task is distributed to task groove X, otherwise, task groove X is joined in the task slot queue;
Telegon is used for the task of initiatively please looking for novelty to scheduling controller according to the resource operating position of place task groove, and the receiving scheduling controller task of distributing, and sends to the tasks carrying device;
The tasks carrying device is used for carrying out the task of distributing.
A kind of execution server comprises two above task grooves, and each task groove all is assigned certain resource, further comprises telegon and tasks carrying device in each task groove;
Telegon is used for the task of initiatively please looking for novelty to scheduling controller according to the resource operating position of place task groove, and the receiving scheduling controller task of distributing, and sends to the tasks carrying device;
The tasks carrying device is used for carrying out the task of distributing.
A kind of Control Server comprises scheduling controller;
Described scheduling controller is used for safeguarding two formations that one is the task waiting list, and another is the task slot queue; When whenever receiving a task X, whether there is the task groove that is fit to Processing tasks X in the slot queue that sets the tasks, if so, then task X is distributed to this task groove, otherwise, task X is joined in the task waiting list; When whenever receiving the task requests that a task groove X initiatively sends, whether exist in the waiting list that sets the tasks to be fit to the task that task groove X processes, if so, then this task is distributed to task groove X, otherwise, task groove X is joined in the task slot queue; Described task requests is that task groove X determines that the resource operating position of self is fit to send after the new task of active request.
As seen, adopt scheme of the present invention, each resource of carrying out server is carried out more fine-grained division, obtain a plurality of task grooves, each task groove all can be executed the task, thereby can maximized utilization respectively carries out the resource of server, has improved resource utilization; In addition, each task groove can initiatively determine whether according to the resource operating position of self to accept new task, and scheduling controller can select to be fit to the task that the task groove is processed when being each task groove allocating task, equally, scheduling controller can select to be fit to process its task groove, so when allocating task, task is processed as early as possible, and can take full advantage of the resource of each task groove, thus not only improved resource utilization, also improved the treatment effeciency of task; Have, scheme of the present invention utilizes telegon to assist scheduling controller to carry out task scheduling, has greatly alleviated the load of scheduling controller, has reached the purpose of load balancing, and has strengthened the extensibility of system again.
Description of drawings
Fig. 1 is that the present invention one carries out the resource schematic diagram that each the task groove in the server is assigned to.
Fig. 2 is the process flow diagram of distributed task dispatching embodiment of the method for the present invention.
Fig. 3 is the composition structural representation of distributed task dispatching system embodiment of the present invention.
Embodiment
For problems of the prior art, the distributed task dispatching scheme among the present invention after a kind of improvement of proposition can improve resource utilization and task treatment effeciency.
Among the present invention, in advance a server in the distributed task dispatching system is defined as Control Server, or be defined as simultaneously Control Server and carry out server, all the other servers all are defined as the execution server, a scheduling controller is set in Control Server, in each execution server, mark off respectively two above task grooves, and be that each task groove distributes required resource.
In actual applications, can carry out certain or certain several resource situation in the server carries out in server and marks off respectively a plurality of task grooves according to each at each, such as, divide according to the CPU number, there are several CPU just to mark off several task grooves, afterwards, according to the task groove number that marks off to other resource, available resources number such as internal memory etc. is divided, division principle can be: each available resources number of carrying out every kind of resource in server is averagely allocated to each task groove in this execution server, namely has:
Wherein, Res
iThe i kind resource that expression one is carried out in the server, Total (Res
i) expression resource Res
iThe available resources number, Task
iExpression task groove i, Count (Taski) represent the task groove number that marks off in this execution server, with ResValue
iDistribute to each the task groove in this execution server.
Fig. 1 is that the present invention one carries out the resource schematic diagram that each the task groove in the server is assigned to.
The task groove possesses the abilities such as active request task, self-equilibrium of stock and tasks carrying.
Scheduling controller is the control center of distributed task dispatching system, and function mainly comprises the task requests of basic task scheduling, task barrel reason and Processing tasks groove etc.
For the consideration to the utilization of resources, scheduling controller and task groove should be the processes of independent operating.Include a background management unit in each server, when the distributed task dispatching system starts, background management unit in the Control Server starts scheduling controller, and be the enough resources of scheduling controller configuration according to the number of servers in the distributed task dispatching system, comprise internal memory and number of threads etc., if Control Server then also needs to start each task groove of dividing simultaneously as carrying out server, afterwards, notice is respectively carried out startup of server; Each background management unit of carrying out in the server starts each task groove of dividing; After finishing above-mentioned processing, each background management unit enters dormant state, withdraws from service.
In the process of distributed task dispatching system operation, scheduling controller can be safeguarded two formations, and one is the task waiting list, another is the task slot queue, wherein, the task waiting list is used for the pending tasks such as preservation, and the task slot queue is used for preserving the task groove of waiting for allocating task.
Fig. 2 is the process flow diagram of distributed task dispatching embodiment of the method for the present invention.As shown in Figure 2, may further comprise the steps:
Step 21: when whenever receiving a task X when (representing any one task with task X), scheduling controller sets the tasks whether there is the task groove that is fit to Processing tasks X in the slot queue, if, then task X is distributed to this task groove, otherwise, task X is joined in the task waiting list.
As previously mentioned, the task slot queue is used for preserving the task groove of waiting for allocating task, in this step, when receiving task X, scheduling controller sets the tasks whether there is the task groove that is fit to Processing tasks X in the slot queue, if, then task X is distributed to this task groove, otherwise, task X is joined in the task waiting list, wait for subsequent treatment.
Wherein, if be accompanied with resource requirement information among receiving of the task X, it is the resource requirement of known task X, then scheduling controller can set the tasks whether there is the task groove of the resource requirement of not using resource can satisfy task X in the slot queue, if existence and number are one, then with its task groove as suitable Processing tasks X, if existence and number are more than two, then therefrom select and do not use resource near the task groove of the resource requirement of task X, with it as the task groove that is fit to Processing tasks X.
If attach resource requirement information among receiving of the task X, namely do not know the resource requirement of task X, then scheduling controller can be selected the task groove that does not use resource maximum from the task slot queue, with its task groove as suitable Processing tasks X.
If the resource requirement of known task X, and the task groove that does not have the resource requirement of not using resource can satisfy task X in the task slot queue, then whether scheduling controller can be determined wherein to exist task X was carried out within the predetermined short period, and when executing the task X, can satisfy the task groove of the resource requirement of task X, if existence and number are one, then with its task groove as suitable Processing tasks X, if existence and number are more than two, then therefrom select the task groove that task X is carried out the soonest, with its task groove as suitable Processing tasks X.
The information such as the not use resource of above-mentioned each task groove can report scheduling controller by the task groove.
Step 22: when whenever receiving the task requests that a task groove X (representing any one task groove with task groove X) initiatively sends, whether scheduling controller sets the tasks to exist in the waiting list and is fit to the task that task groove X processes, if, then this task is distributed to task groove X, otherwise, task groove X is joined in the task slot queue; Described task requests is that task groove X determines that the resource operating position of self is fit to send after the new task of active request.
In actual applications, each task groove can periodically calculate the resource scoring of reflection own resource operating position, and result of calculation and the predefined threshold value of asking compared, if result of calculation be greater than asking threshold value, before then determining whether to the excessively new task of scheduling controller request but unallocated to new task, if, then do not deal with, otherwise, please looking for novelty to scheduling controller of task, and self resource operating position reported scheduling controller.
Still take task groove X as example, its in the following manner computational resource scoring:
ResScore＝∑Res
i*Ratio
i；(2)
Wherein, the resource scoring that ResScore namely represents to calculate, Res
iThe operating position that represents i kind resource, Ratio
iRepresent weight corresponding to i kind resource.
If ResScore is greater than the predefined threshold value of asking, then task groove X can be to the scheduling controller request task, but, might before the ResScore that goes out of some computation of Period also greater than asking threshold value, thereby to the scheduling controller request task, but scheduling controller does not have suitable task to distribute to task groove X, like this, just there is no need again repetitive requests, therefore, task groove X is determining ResScore greater than after can asking threshold value, before needing at first to determine whether to the excessively new task of scheduling controller request but unallocated to new task, if, then do not deal with, otherwise, please looking for novelty to scheduling controller of task, and, self resource operating position is reported scheduling controller.
After scheduling controller receives the task requests that task groove X sends, whether exist in the waiting list that sets the tasks to be fit to the task that task groove X processes, if exist, then this task is distributed to task groove X, otherwise, task groove X is joined in the task slot queue.Which type of task can with reference to above stated specification, repeat no more for the task of being fit to task groove X processing herein.
After task groove X received the task of distributing, at first definite task of distributing could be carried out immediately, if so, then carry out immediately, otherwise, the task of distributing is carried out buffer memory, after current task is finished, carry out again the task of buffer memory.
In actual applications, if necessary, recyclable having distributed to the task groove but not being performed of the task of scheduling controller, and reclaiming of task processed as newly receiving of task, namely process according to mode shown in the step 21.
Need to prove, step 21 and 22 when carrying out, do not have sequencing minute, executed in parallel normally, expression mode shown in Figure 2 only in order to express easily.
Below by concrete example, scheme of the present invention is further specified.
Suppose to comprise altogether in certain distributed task dispatching system 3 servers, wherein 1 server possesses CPU, the 4G internal memory of 4 3GHz, other 2 servers possess CPU, the 3G internal memory of 2 2GHz, so, can will possess the server of CPU, 4G internal memory of 4 3GHz as Control Server, and configure a scheduling controller, 3 task grooves, and be that scheduling controller distributes the 1G internal memory, for each task groove distributes the 600M internal memory, remaining internal memory can be kept for other application beyond the scheme of the present invention; All as carrying out server, each configures 3 task grooves to remaining 2 servers, and is that each task groove distributes the 700M internal memory, and same, remaining internal memory can be kept for other application beyond the scheme of the present invention; Like this, obtain altogether 9 task grooves, i.e. 6＜700M, 2G〉task groove and 3＜600M, 3G〉the task groove.
After the distributed task dispatching system started, each task groove was all to the scheduling controller request task, owing to also not having task to carry out this moment, so all task grooves all enter the task slot queue, and along with the continuous arrival of task, each task groove all is assigned to task.
Follow-up, each task groove take task groove X as example, periodically calculates the resource scoring of self, as:
Resource scoring=CPU idleness * 40%+ free memory rate * 80%+ (maximum configured Thread Count/reality is used Thread Count) * 5%; (3)
Wherein 40%, 80% and 5% be weight, CPU idleness, free memory rate and the actual thread number average that uses refer to mean value in the one-period.
Suppose that the CPU idleness is 40%, the free memory rate also is 40%, it is 4 that maximum configured Thread Count/reality is used Thread Count, can calculate so resource scoring is 0.68, greater than can asking threshold value 0.5, and sends task requests to scheduling controller before not existing but is not assigned to the situation of task, then task groove X-direction scheduling controller sends task requests, and report self resource operating position, as＜700M, 40%; 2G, 40% 〉, in addition, also the overall tasks implementation progress of self can be reported scheduling controller, wherein, carry out at the same time a plurality of tasks if task groove X is current, the overall tasks implementation progress refers to the average implementation progress of these a plurality of tasks so.
Scheduling controller receives from after the task requests of task groove X and the resource operating position that reports and the overall tasks implementation progress, according to＜700M, 40%; 2G, 40%〉calculate, the internal memory that namely distributes with task groove X multiply by free memory rate 40%, obtains untapped internal memory 280M, uses the CPU2G that distributes to multiply by CPU idleness 40%, obtains untapped CPU 0.8G, namely has＜280M; 0.8G.By searching the task waiting list, the task that it is 100M that discovery wherein has a memory requirements can be distributed to task groove X, so, this task is taken out from the task waiting list, distributes to task groove X.
For a memory requirements that newly the receives task X that is 700M, after if scheduling controller has been checked the resource operating position of each the task groove in the task slot queue, find not use resource can satisfy the task groove of the resource requirement of task X, but the overall tasks implementation progress that a task groove is arranged is 90%, greater than predetermined threshold value 70%, and this task groove distribute in save as 700M, this shows if task X is distributed to this task groove, task X will be processed very soon so, and this task groove can satisfy the resource requirement of task X then, therefore, task X can be distributed to this task groove.
In addition, be further to improve resource utilization, in the scheme of the present invention, also can adjust each task groove number of carrying out in server according to resource operating position etc., described adjustment comprises increases and reduces (merging), is described below respectively.
1) reduces
Specifically, each task groove X periodically calculates the resource scoring of self, and result of calculation and the predefined minimum threshold value that limits compared, if the result of calculation in a continuous N cycle all limits threshold value less than minimum, then can be after executing all tasks of having distributed, the oneself closes, releasing resource, and the notice scheduling controller, M is the positive integer greater than 1; Scheduling controller is distributed to the resource that discharges other task groove in the execution server at task groove X place, and so, the task groove number in the execution server at task groove X place has just reduced one.
As a rule, the task groove that only is in full load condition just need to increase resource, and therefore, task groove X can send resource release request to scheduling controller first after the result of calculation of determining a continuous N cycle all limits threshold value less than minimum; After scheduling controller receives resource release request from task groove X, set the tasks and whether have the task groove that is in full load condition in the execution server at groove X place, if, then notify task groove X releasing resource, and no longer be task groove X allocating task, otherwise, do not allow task groove X releasing resource; Follow-up, the resource that scheduling controller discharges task groove X is distributed to the task groove that is in full load condition, is a plurality of if be in the task groove of full load condition, can therefrom select at random one.
2) increase
Each task groove X periodically calculates the resource scoring of self, and result of calculation and the predefined minimum threshold value that limits compared, if the result of calculation in a continuous N cycle is all less than minimum restriction threshold value, then release portion resource, and the notice scheduling controller, M is the positive integer greater than 1; Scheduling controller creates a new task groove in the execution server at task groove X place, and the resource that discharges is distributed to the task groove of new establishment, and so, the task groove number in the execution server at task groove X place has just increased by one.
In actual applications, can limit carrying out the task groove number that allows in the server to exist, a maximum task groove limit number namely is set, like this, task groove X can send resource release request to scheduling controller first before releasing resource; Correspondingly, whether scheduling controller sets the tasks task groove number in the execution server at groove X place less than maximum task groove limit number, if so, then notifies task groove X releasing resource, otherwise, do not allow task groove X releasing resource.
The stock number that task groove X discharges can be decided according to the actual requirements, such as discharging half.
So far, namely finished introduction about the inventive method embodiment.
Based on above-mentioned introduction, Fig. 3 is the composition structural representation of distributed task dispatching system embodiment of the present invention.As shown in Figure 3, comprising: two above servers, one of them is Control Server, or is Control Server and execution server simultaneously, all the other are the execution server; Comprise a scheduling controller in the Control Server, each is carried out in the server and includes two above task grooves, and each task groove all is assigned certain resource, further comprises telegon and tasks carrying device in each task groove.
Scheduling controller is used for safeguarding two formations that one is the task waiting list, and another is the task slot queue; When whenever receiving a task X, whether there is the task groove that is fit to Processing tasks X in the slot queue that sets the tasks, if so, then task X is distributed to this task groove, otherwise, task X is joined in the task waiting list; When whenever receiving the task requests that a task groove X initiatively sends, whether exist in the waiting list that sets the tasks to be fit to the task that task groove X processes, if so, then this task is distributed to task groove X, otherwise, task groove X is joined in the task slot queue;
Telegon is used for the task of initiatively please looking for novelty to scheduling controller according to the resource operating position of place task groove, and the receiving scheduling controller task of distributing, and sends to the tasks carrying device;
The tasks carrying device is used for carrying out the task of distributing.
Wherein, telegon can periodically calculate the resource scoring of reflection place task groove resource operating position, and result of calculation and the predefined threshold value of asking compared, if result of calculation be greater than asking threshold value, before then further determining whether to the excessively new task of scheduling controller request but unallocated to new task, if, then do not deal with, otherwise, please looking for novelty to scheduling controller of task, and self resource operating position reported scheduling controller.
In addition, if the resource requirement of known task X, then scheduling controller sets the tasks whether there is the task groove of the resource requirement of not using resource can satisfy task X in the slot queue, if existence and number are one, then with its task groove as suitable Processing tasks X, if exist and number is more than two, then therefrom select and do not use resource near the task groove of the resource requirement of task X, with its task groove as suitable Processing tasks X; If do not know the resource requirement of task X, then scheduling controller is selected the task groove that does not use resource maximum from the task slot queue, with its task groove as suitable Processing tasks X; If the resource requirement of known task X, and the task groove that does not have the resource requirement of not using resource can satisfy task X in the task slot queue, then whether scheduling controller is determined wherein to exist task X was carried out within the predetermined short period, and can satisfy the task groove of the resource requirement of task X when executing the task X; If existence and number are one, then with its task groove as suitable Processing tasks X, if existence and number are more than two, then therefrom select the task groove that task X is carried out the soonest, with its task groove as suitable Processing tasks X.
Above-mentioned telegon can be further used for, and determines that distributing of task could carry out immediately, if so, then sends to the tasks carrying device, otherwise, the task of distributing is carried out buffer memory, after current task is finished, the task of buffer memory is sent to the tasks carrying device; Correspondingly, scheduling controller can be further used for, and reclaims having distributed to the task groove but not being performed of task.
In addition, above-mentioned telegon also can be further used for, periodically calculate the resource scoring of place task groove, and result of calculation and the predefined minimum threshold value that limits compared, if the result of calculation in a continuous N cycle is all less than minimum restriction threshold value, then after the task actuator executes all tasks of having distributed, close place task groove, releasing resource, and notice scheduling controller, M is the positive integer greater than 1; Correspondingly, scheduling controller is further used for, and the resource that discharges is distributed to other task groove in the execution server at task groove place of releasing resource.
Above-mentioned telegon also can be further used for, periodically calculate the resource scoring of place task groove, and result of calculation and the predefined minimum threshold value that limits compared, if the result of calculation in a continuous N cycle all limits threshold value less than minimum, release portion resource then, and the notice scheduling controller, M is the positive integer greater than 1; Correspondingly, scheduling controller is further used for, and creates a new task groove in the execution server at the task groove place of release portion resource, and the resource that discharges is distributed to the task groove of new establishment.
The specific works flow process of system shown in Figure 3 embodiment please refer to the respective description among the said method embodiment, repeats no more herein.
Need to prove, each threshold value that relates in the various embodiments described above and the value of weight all can be decided according to the actual requirements.
In a word, adopt scheme of the present invention, each resource of carrying out server is carried out more fine-grained division, obtain a plurality of task grooves, each task groove all can be executed the task, thereby can maximized utilization respectively carries out the resource of server, has improved resource utilization; In addition, each task groove can initiatively determine whether according to the resource operating position of self to accept new task, and scheduling controller can select to be fit to the task that the task groove is processed when being each task groove allocating task, equally, scheduling controller can select to be fit to process its task groove, so when allocating task, task is processed as early as possible, and can take full advantage of the resource of each task groove, thus not only improved resource utilization, also improved the treatment effeciency of task; Have, scheme of the present invention utilizes telegon to assist scheduling controller to carry out task scheduling, has greatly alleviated the load of scheduling controller, has reached the purpose of load balancing, and has strengthened the extensibility of system again.
The above only is preferred embodiment of the present invention, and is in order to limit the present invention, within the spirit and principles in the present invention not all, any modification of making, is equal to replacement, improvement etc., all should be included within the scope of protection of the invention.
Claims (25)
1. distributed task dispatching method, it is characterized in that, a Control Server is set in the distributed task dispatching system and carries out server more than one, a scheduling controller is set in Control Server, in each execution server, mark off respectively two above task grooves, and be that each task groove distributes required resource; When the operation of described distributed task dispatching system, the method comprises:
Scheduling controller is safeguarded two formations, and one is the task waiting list, and another is the task slot queue;
When whenever receiving a task X, whether scheduling controller sets the tasks exists the task groove that is fit to Processing tasks X in the slot queue, if so, then task X is distributed to this task groove, otherwise, task X is joined in the task waiting list;
When whenever receiving the task requests that a task groove X initiatively sends, whether scheduling controller sets the tasks to exist in the waiting list and is fit to the task that task groove X processes, and if so, then this task is distributed to task groove X, otherwise, task groove X is joined in the task slot queue; Described task requests is that task groove X determines that the own resource operating position is fit to send after the new task of active request.
2. method according to claim 1 is characterized in that, the method further comprises:
Each task groove periodically calculates the resource scoring of self, and result of calculation and the predefined threshold value of asking compared, if result of calculation is greater than asking threshold value, before then determining whether to the excessively new task of scheduling controller request but unallocated to new task, if so, then do not deal with, otherwise, please looking for novelty to scheduling controller of task, and self resource operating position reported scheduling controller.
3. method according to claim 2 is characterized in that, whether described scheduling controller sets the tasks exists in the slot queue task groove that is fit to Processing tasks X to comprise:
If the resource requirement of known task X, then set the tasks and whether have the task groove of the resource requirement of not using resource can satisfy task X in the slot queue, if existence and number are one, then with its task groove as suitable Processing tasks X, if existence and number are more than two, then therefrom select and do not use resource near the task groove of the resource requirement of task X, with it as the task groove that is fit to Processing tasks X;
If do not know the resource requirement of task X, then from the task slot queue, select the task groove that does not use resource maximum, with its task groove as suitable Processing tasks X.
4. method according to claim 3 is characterized in that, the method further comprises:
If the resource requirement of known task X, and the task groove that does not have the resource requirement of not using resource can satisfy task X in the task slot queue, then determine wherein whether to exist task X was carried out within the predetermined short period, and when executing the task X, can satisfy the task groove of the resource requirement of task X; If existence and number are one, then with its task groove as suitable Processing tasks X, if existence and number are more than two, then therefrom select the task groove that task X is carried out the soonest, with its task groove as suitable Processing tasks X.
5. method according to claim 1 is characterized in that, the method further comprises:
After each task groove received the task of distributing, definite task of distributing could be carried out immediately, if so, then carry out immediately, otherwise, the task of distributing is carried out buffer memory, after current task is finished, carry out the task of buffer memory.
6. method according to claim 5 is characterized in that, the method further comprises: scheduling controller reclaims having distributed to the task groove but not being performed of task.
7. according to claim 1,5 or 6 described methods, it is characterized in that, the method further comprises:
Each task groove periodically calculates the resource scoring of self, and result of calculation and the predefined minimum threshold value that limits compared, if the result of calculation in a continuous N cycle all limits threshold value less than minimum, then after executing all tasks of having distributed, the oneself closes, releasing resource, and notice scheduling controller; M is the positive integer greater than 1;
Scheduling controller is distributed to the resource that discharges other task groove in the execution server at task groove place of releasing resource.
8. method according to claim 7 is characterized in that, determines that the result of calculation in a continuous N cycle all less than after the minimum restriction threshold value, further comprises:
Send resource release request to scheduling controller; Whether have the task groove that is in full load condition in the execution server at the task groove place of the definite transmission of scheduling controller resource release request, if so, then notice sends the task groove releasing resource of resource release request;
Described other task groove is: the task groove that is in full load condition.
9. according to claim 1,5 or 6 described methods, it is characterized in that, the method further comprises:
Each task groove periodically calculates the resource scoring of self, and result of calculation and the predefined minimum threshold value that limits are compared, if the result of calculation in a continuous N cycle all limits threshold value less than minimum, and then release portion resource, and notice scheduling controller; M is the positive integer greater than 1;
Scheduling controller creates a new task groove in the execution server at the task groove place of release portion resource, and the resource that discharges is distributed to the task groove of new establishment.
10. method according to claim 9 is characterized in that, before the described release portion resource, further comprises:
Send resource release request to scheduling controller;
Whether the task groove number in the execution server at the task groove place of the definite transmission of scheduling controller resource release request is less than predefined maximum task groove limit number, and if so, then notice sends the task groove releasing resource of resource release request.
11. a distributed task dispatching system is characterized in that, comprising: two above servers, one of them is Control Server, or is Control Server and execution server simultaneously, all the other are the execution server; Comprise a scheduling controller in the Control Server, each is carried out in the server and includes two above task grooves, and each task groove all is assigned certain resource, further comprises telegon and tasks carrying device in each task groove;
Scheduling controller is used for safeguarding two formations that one is the task waiting list, and another is the task slot queue; When whenever receiving a task X, whether there is the task groove that is fit to Processing tasks X in the slot queue that sets the tasks, if so, then task X is distributed to this task groove, otherwise, task X is joined in the task waiting list; When whenever receiving the task requests that a task groove X initiatively sends, whether exist in the waiting list that sets the tasks to be fit to the task that task groove X processes, if so, then this task is distributed to task groove X, otherwise, task groove X is joined in the task slot queue;
Telegon is used for the task of initiatively please looking for novelty to scheduling controller according to the resource operating position of place task groove, and the receiving scheduling controller task of distributing, and sends to the tasks carrying device;
The tasks carrying device is used for carrying out the task of distributing.
12. system according to claim 11, it is characterized in that, described telegon periodically calculates the resource scoring of place task groove, and result of calculation and the predefined threshold value of asking compared, if result of calculation is greater than asking threshold value, before then further determining whether to the excessively new task of scheduling controller request but unallocated to new task, if, then do not deal with, otherwise, please looking for novelty to scheduling controller of task, and self resource operating position reported scheduling controller.
13. system according to claim 12 is characterized in that,
If the resource requirement of known task X, then scheduling controller sets the tasks whether there is the task groove of the resource requirement of not using resource can satisfy task X in the slot queue, if existence and number are one, then with its task groove as suitable Processing tasks X, if existence and number are more than two, then therefrom select and do not use resource near the task groove of the resource requirement of task X, with it as the task groove that is fit to Processing tasks X;
If do not know the resource requirement of task X, then scheduling controller is selected the task groove that does not use resource maximum from the task slot queue, with its task groove as suitable Processing tasks X;
If the resource requirement of known task X, and the task groove that does not have the resource requirement of not using resource can satisfy task X in the task slot queue, then whether scheduling controller is determined wherein to exist task X was carried out within the predetermined short period, and can satisfy the task groove of the resource requirement of task X when executing the task X; If existence and number are one, then with its task groove as suitable Processing tasks X, if existence and number are more than two, then therefrom select the task groove that task X is carried out the soonest, with its task groove as suitable Processing tasks X.
14. system according to claim 11, it is characterized in that, described telegon is further used for, definite task of distributing could be carried out immediately, if so, then sends to the tasks carrying device, otherwise, the task of distributing is carried out buffer memory, after current task is finished, the task of buffer memory is sent to the tasks carrying device.
15. system according to claim 14 is characterized in that, described scheduling controller is further used for, and reclaims having distributed to the task groove but not being performed of task.
16. according to claim 11,14 or 15 described systems, it is characterized in that,
Described telegon is further used for, periodically calculate the resource scoring of place task groove, and result of calculation and the predefined minimum threshold value that limits compared, if the result of calculation in a continuous N cycle all limits threshold value less than minimum, then after the task actuator executes all tasks of having distributed, close place task groove, releasing resource, and notice scheduling controller; M is the positive integer greater than 1;
Described scheduling controller is further used for, and the resource that discharges is distributed to other task groove in the execution server at task groove place of releasing resource.
17. according to claim 11,14 or 15 described systems, it is characterized in that,
Described telegon is further used for, periodically calculate the resource scoring of place task groove, and result of calculation and the predefined minimum threshold value that limits are compared, if the result of calculation in a continuous N cycle all limits threshold value less than minimum, then release portion resource, and notice scheduling controller; M is the positive integer greater than 1;
Described scheduling controller is further used for, and creates a new task groove in the execution server at the task groove place of release portion resource, and the resource that discharges is distributed to the task groove of new establishment.
18. carry out server for one kind, it is characterized in that, comprise two above task grooves, each task groove all is assigned certain resource, further comprises telegon and tasks carrying device in each task groove;
Telegon is used for the task of initiatively please looking for novelty to scheduling controller according to the resource operating position of place task groove, and the receiving scheduling controller task of distributing, and sends to the tasks carrying device;
The tasks carrying device is used for carrying out the task of distributing.
19. execution server according to claim 18, it is characterized in that, described telegon periodically calculates the resource scoring of place task groove, and result of calculation and the predefined threshold value of asking compared, if result of calculation is greater than asking threshold value, before then further determining whether to the excessively new task of scheduling controller request but unallocated to new task, if, then do not deal with, otherwise, please looking for novelty to scheduling controller of task, and self resource operating position reported scheduling controller.
20. execution server according to claim 18, it is characterized in that, described telegon is further used for, definite task of distributing could be carried out immediately, if so, then sends to the tasks carrying device, otherwise, the task of distributing is carried out buffer memory, after current task is finished, the task of buffer memory is sent to the tasks carrying device.
21. according to claim 18 or 20 described execution servers, it is characterized in that, described telegon is further used for, periodically calculate the resource scoring of place task groove, and result of calculation and the predefined minimum threshold value that limits compared, if the result of calculation in a continuous N cycle all limits threshold value less than minimum, then after the task actuator executes all tasks of having distributed, close place task groove, releasing resource, and the notice scheduling controller, so that scheduling controller is distributed to other task groove with the resource that discharges; M is the positive integer greater than 1.
22. according to claim 18 or 20 described execution servers, it is characterized in that, described telegon is further used for, periodically calculate the resource scoring of place task groove, and result of calculation and the predefined minimum threshold value that limits are compared, if the result of calculation in a continuous N cycle all limits threshold value less than minimum, release portion resource then, and the notice scheduling controller, so that scheduling controller creates a new task groove, and the resource that discharges is distributed to the task groove of new establishment; M is the positive integer greater than 1.
23. a Control Server is characterized in that, comprises scheduling controller;
Described scheduling controller is used for safeguarding two formations that one is the task waiting list, and another is the task slot queue; When whenever receiving a task X, whether there is the task groove that is fit to Processing tasks X in the slot queue that sets the tasks, if so, then task X is distributed to this task groove, otherwise, task X is joined in the task waiting list; When whenever receiving the task requests that a task groove X initiatively sends, whether exist in the waiting list that sets the tasks to be fit to the task that task groove X processes, if so, then this task is distributed to task groove X, otherwise, task groove X is joined in the task slot queue; Described task requests is that task groove X determines that the resource operating position of self is fit to send after the new task of active request.
24. Control Server according to claim 23 is characterized in that,
If the resource requirement of known task X, then scheduling controller sets the tasks whether there is the task groove of the resource requirement of not using resource can satisfy task X in the slot queue, if existence and number are one, then with its task groove as suitable Processing tasks X, if existence and number are more than two, then therefrom select and do not use resource near the task groove of the resource requirement of task X, with it as the task groove that is fit to Processing tasks X;
If do not know the resource requirement of task X, then scheduling controller is selected the task groove that does not use resource maximum from the task slot queue, with its task groove as suitable Processing tasks X;
If the resource requirement of known task X, and the task groove that does not have the resource requirement of not using resource can satisfy task X in the task slot queue, then whether scheduling controller is determined wherein to exist task X was carried out within the predetermined short period, and can satisfy the task groove of the resource requirement of task X when executing the task X; If existence and number are one, then with its task groove as suitable Processing tasks X, if existence and number are more than two, then therefrom select the task groove that task X is carried out the soonest, with its task groove as suitable Processing tasks X.
25. according to claim 23 or 24 described Control Servers, it is characterized in that, described scheduling controller is further used for, and reclaims having distributed to the task groove but not being performed of task.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110214428.2A CN102902587B (en) | 2011-07-28 | 2011-07-28 | Distributed task dispatching mthods, systems and devices |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110214428.2A CN102902587B (en) | 2011-07-28 | 2011-07-28 | Distributed task dispatching mthods, systems and devices |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102902587A true CN102902587A (en) | 2013-01-30 |
| CN102902587B CN102902587B (en) | 2015-10-07 |
Family
ID=47574833
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110214428.2A Active CN102902587B (en) | 2011-07-28 | 2011-07-28 | Distributed task dispatching mthods, systems and devices |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102902587B (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014146279A1 (en) * | 2013-03-21 | 2014-09-25 | Telefonaktiebolaget L M Ericsson (Publ) | Method and device for scheduling communication schedulable unit |
| CN104734923A (en) * | 2013-12-18 | 2015-06-24 | 青岛海尔空调器有限总公司 | Execution control method and system of application task and information server |
| CN105630601A (en) * | 2014-11-03 | 2016-06-01 | 阿里巴巴集团控股有限公司 | Resource allocation method and system based on real-time computing |
| CN105718317A (en) * | 2016-01-15 | 2016-06-29 | 浪潮(北京)电子信息产业有限公司 | Task scheduling method and task scheduling device |
| CN106020988A (en) * | 2016-06-03 | 2016-10-12 | 北京邮电大学 | Off-line task scheduling method and device for intelligent video monitoring system |
| CN106302587A (en) * | 2015-05-26 | 2017-01-04 | 阿里巴巴集团控股有限公司 | The dispatching method of a kind of internet information resource and device |
| CN106557366A (en) * | 2015-09-28 | 2017-04-05 | 阿里巴巴集团控股有限公司 | Task distribution method, apparatus and system |
| CN107092999A (en) * | 2016-11-08 | 2017-08-25 | 北京小度信息科技有限公司 | Task processing method and device |
| CN107239328A (en) * | 2016-03-29 | 2017-10-10 | 上海普兰金融服务有限公司 | Method for allocating tasks and device |
| CN107590002A (en) * | 2017-09-15 | 2018-01-16 | 东软集团股份有限公司 | Method for allocating tasks, device, storage medium, equipment and distributed task scheduling system |
| CN107704320A (en) * | 2017-05-12 | 2018-02-16 | 贵州白山云科技有限公司 | The method for allocating tasks and system of a kind of distributed system |
| CN108182111A (en) * | 2018-01-23 | 2018-06-19 | 百度在线网络技术（北京）有限公司 | Task scheduling system, method and apparatus |
| CN108255609A (en) * | 2018-01-12 | 2018-07-06 | 瑞银国际物流（江苏）有限公司 | A kind of Web VR distributed computing methods based on device chip |
| CN108647090A (en) * | 2018-04-03 | 2018-10-12 | 联想(北京)有限公司 | A kind of resource allocation methods and device, server cluster |
| CN108737462A (en) * | 2017-04-17 | 2018-11-02 | 华东师范大学 | A kind of cloud computation data center method for scheduling task based on graph theory |
| CN109101340A (en) * | 2018-09-18 | 2018-12-28 | 武汉斗鱼网络科技有限公司 | Dynamic distribution method, device, equipment and the storage medium of barrage message |
| CN110046034A (en) * | 2018-01-15 | 2019-07-23 | 北京国双科技有限公司 | Task acquisition methods and device |
| CN111090507A (en) * | 2019-11-25 | 2020-05-01 | 南京航空航天大学 | Task scheduling method based on cloud edge fusion server network architecture and application |
| CN111679900A (en) * | 2020-06-15 | 2020-09-18 | 杭州海康威视数字技术股份有限公司 | Task processing method and device |
| CN111694672A (en) * | 2020-06-12 | 2020-09-22 | 北京字节跳动网络技术有限公司 | Resource allocation method, task submission method, device, electronic equipment and medium |
| CN112785323A (en) * | 2019-11-07 | 2021-05-11 | 北京沃东天骏信息技术有限公司 | Resource allocation method and device and electronic equipment |
| CN113419826A (en) * | 2021-05-11 | 2021-09-21 | 北京天云融创软件技术有限公司 | Flexible resource limiting method for task scheduling |
| WO2024066342A1 (en) * | 2022-09-26 | 2024-04-04 | 京东科技信息技术有限公司 | Task processing method and apparatus, electronic device, and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080120621A1 (en) * | 2006-11-17 | 2008-05-22 | Fujitsu Limited | Resource management apparatus and radio network controller |
| CN101236513A (en) * | 2007-01-30 | 2008-08-06 | 阿里巴巴公司 | Distributed task system and distributed task management process |
| CN101957780A (en) * | 2010-08-17 | 2011-01-26 | 中国电子科技集团公司第二十八研究所 | Resource state information-based grid task scheduling processor and grid task scheduling processing method |
-
2011
- 2011-07-28 CN CN201110214428.2A patent/CN102902587B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080120621A1 (en) * | 2006-11-17 | 2008-05-22 | Fujitsu Limited | Resource management apparatus and radio network controller |
| CN101236513A (en) * | 2007-01-30 | 2008-08-06 | 阿里巴巴公司 | Distributed task system and distributed task management process |
| CN101957780A (en) * | 2010-08-17 | 2011-01-26 | 中国电子科技集团公司第二十八研究所 | Resource state information-based grid task scheduling processor and grid task scheduling processing method |
Cited By (37)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014146279A1 (en) * | 2013-03-21 | 2014-09-25 | Telefonaktiebolaget L M Ericsson (Publ) | Method and device for scheduling communication schedulable unit |
| CN105284084A (en) * | 2013-03-21 | 2016-01-27 | 爱立信(中国)通信有限公司 | Method and device for scheduling communication schedulable unit |
| US9471372B2 (en) | 2013-03-21 | 2016-10-18 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and device for scheduling communication schedulable unit |
| CN105284084B (en) * | 2013-03-21 | 2018-11-06 | 爱立信(中国)通信有限公司 | Method and apparatus for dispatching communication schedulable unit |
| CN104734923A (en) * | 2013-12-18 | 2015-06-24 | 青岛海尔空调器有限总公司 | Execution control method and system of application task and information server |
| CN104734923B (en) * | 2013-12-18 | 2018-12-28 | 青岛海尔空调器有限总公司 | The execution control method and system of application task, message server |
| CN105630601A (en) * | 2014-11-03 | 2016-06-01 | 阿里巴巴集团控股有限公司 | Resource allocation method and system based on real-time computing |
| CN106302587B (en) * | 2015-05-26 | 2019-07-30 | 阿里巴巴集团控股有限公司 | A kind of dispatching method and device of internet information resource |
| CN106302587A (en) * | 2015-05-26 | 2017-01-04 | 阿里巴巴集团控股有限公司 | The dispatching method of a kind of internet information resource and device |
| WO2017054650A1 (en) * | 2015-09-28 | 2017-04-06 | 阿里巴巴集团控股有限公司 | Task distribution method, device and system |
| CN106557366A (en) * | 2015-09-28 | 2017-04-05 | 阿里巴巴集团控股有限公司 | Task distribution method, apparatus and system |
| CN106557366B (en) * | 2015-09-28 | 2020-09-08 | 阿里巴巴集团控股有限公司 | Task distribution method, device and system |
| CN105718317A (en) * | 2016-01-15 | 2016-06-29 | 浪潮(北京)电子信息产业有限公司 | Task scheduling method and task scheduling device |
| CN107239328A (en) * | 2016-03-29 | 2017-10-10 | 上海普兰金融服务有限公司 | Method for allocating tasks and device |
| CN106020988B (en) * | 2016-06-03 | 2019-03-15 | 北京邮电大学 | A kind of offline method for scheduling task of intelligent video monitoring system and device |
| CN106020988A (en) * | 2016-06-03 | 2016-10-12 | 北京邮电大学 | Off-line task scheduling method and device for intelligent video monitoring system |
| CN107092999A (en) * | 2016-11-08 | 2017-08-25 | 北京小度信息科技有限公司 | Task processing method and device |
| CN108737462A (en) * | 2017-04-17 | 2018-11-02 | 华东师范大学 | A kind of cloud computation data center method for scheduling task based on graph theory |
| CN107704320A (en) * | 2017-05-12 | 2018-02-16 | 贵州白山云科技有限公司 | The method for allocating tasks and system of a kind of distributed system |
| WO2018205890A1 (en) * | 2017-05-12 | 2018-11-15 | 贵州白山云科技有限公司 | Task assignment method and system of distributed system, computer readable storage medium and computer device therefor |
| CN107590002A (en) * | 2017-09-15 | 2018-01-16 | 东软集团股份有限公司 | Method for allocating tasks, device, storage medium, equipment and distributed task scheduling system |
| CN108255609A (en) * | 2018-01-12 | 2018-07-06 | 瑞银国际物流（江苏）有限公司 | A kind of Web VR distributed computing methods based on device chip |
| CN110046034A (en) * | 2018-01-15 | 2019-07-23 | 北京国双科技有限公司 | Task acquisition methods and device |
| CN108182111B (en) * | 2018-01-23 | 2022-07-15 | 百度在线网络技术（北京）有限公司 | Task scheduling system, method and device |
| CN108182111A (en) * | 2018-01-23 | 2018-06-19 | 百度在线网络技术（北京）有限公司 | Task scheduling system, method and apparatus |
| CN108647090A (en) * | 2018-04-03 | 2018-10-12 | 联想(北京)有限公司 | A kind of resource allocation methods and device, server cluster |
| CN109101340B (en) * | 2018-09-18 | 2021-11-26 | 武汉斗鱼网络科技有限公司 | Dynamic distribution method, device, equipment and storage medium of barrage message |
| CN109101340A (en) * | 2018-09-18 | 2018-12-28 | 武汉斗鱼网络科技有限公司 | Dynamic distribution method, device, equipment and the storage medium of barrage message |
| CN112785323A (en) * | 2019-11-07 | 2021-05-11 | 北京沃东天骏信息技术有限公司 | Resource allocation method and device and electronic equipment |
| CN111090507A (en) * | 2019-11-25 | 2020-05-01 | 南京航空航天大学 | Task scheduling method based on cloud edge fusion server network architecture and application |
| CN111090507B (en) * | 2019-11-25 | 2023-06-09 | 南京航空航天大学 | Task scheduling method and application based on cloud edge fusion server network architecture |
| CN111694672A (en) * | 2020-06-12 | 2020-09-22 | 北京字节跳动网络技术有限公司 | Resource allocation method, task submission method, device, electronic equipment and medium |
| CN111694672B (en) * | 2020-06-12 | 2023-04-25 | 抖音视界有限公司 | Resource allocation method, task submission method, device, electronic equipment and medium |
| CN111679900A (en) * | 2020-06-15 | 2020-09-18 | 杭州海康威视数字技术股份有限公司 | Task processing method and device |
| CN111679900B (en) * | 2020-06-15 | 2023-10-31 | 杭州海康威视数字技术股份有限公司 | Task processing method and device |
| CN113419826A (en) * | 2021-05-11 | 2021-09-21 | 北京天云融创软件技术有限公司 | Flexible resource limiting method for task scheduling |
| WO2024066342A1 (en) * | 2022-09-26 | 2024-04-04 | 京东科技信息技术有限公司 | Task processing method and apparatus, electronic device, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102902587B (en) | 2015-10-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102902587A (en) | Distribution type task scheduling method, distribution type task scheduling system and distribution type task scheduling device | |
| CN103793272B (en) | Periodical task scheduling method and periodical task scheduling system | |
| Levin et al. | DP-FAIR: A simple model for understanding optimal multiprocessor scheduling | |
| CN102063336B (en) | Distributed computing multiple application function asynchronous concurrent scheduling method | |
| CN102387173B (en) | MapReduce system and method and device for scheduling tasks thereof | |
| CN101452404B (en) | Task scheduling apparatus and method for embedded operating system | |
| CN102096599A (en) | Multi-queue task scheduling method and related system and equipment | |
| CN103268247B (en) | Method and device for executing task and adjusting number of remaining threads in thread pool | |
| CN101984414B (en) | Method and device for scheduling central processing unit (CPU) resource | |
| CN112269641A (en) | Scheduling method, scheduling device, electronic equipment and storage medium | |
| CN101145112A (en) | Real-time system task scheduling method | |
| CN103179048A (en) | Method and system for changing main machine quality of service (QoS) strategies of cloud data center | |
| CN112783659A (en) | Resource allocation method and device, computer equipment and storage medium | |
| CN114371926B (en) | Refined resource allocation method and device, electronic equipment and medium | |
| CN103455375A (en) | Load-monitoring-based hybrid scheduling method under Hadoop cloud platform | |
| CN109729113A (en) | Manage method, server system and the computer program product of dedicated processes resource | |
| CN109189581B (en) | Job scheduling method and device | |
| CN102184124B (en) | Task scheduling method and system | |
| CN112882827A (en) | Method, electronic device and computer program product for load balancing | |
| Ghouma et al. | Context aware resource allocation and scheduling for mobile cloud | |
| Dahal et al. | Scheduling in multiprocessor system using genetic algorithms | |
| CN111210286A (en) | Tax control server-based efficient invoice issuing method and system | |
| CN114138453B (en) | Resource optimization allocation method and system suitable for edge computing environment | |
| Sirohi et al. | Improvised round robin (CPU) scheduling algorithm | |
| US20190340034A1 (en) | Device and method for processing jobs |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20160315 Address after: West high tech Zone Fucheng Road in Chengdu city of Sichuan province 610000 399 No. 6 Building 1 unit 12 floor No. 3 Patentee after: MIGU MUSIC CO., LTD. Address before: 610041 No. 10 Peng Da Road, hi tech Zone, Sichuan, Chengdu Patentee before: China Mobile Communication Group Sichuan Co., Ltd. |